LO
AI implementation lead — how we know
The page itself gives the judgements. This one gives what they rest on: which technologies bear on the work, how the estimate moved since language models reached the public, and the method behind both.
Which technologies matter here#
Four separate signals. They are deliberately not added together — a job exposed to two technologies is not twice as exposed.
How it got here#
The index is not a static number. This is where it would have sat at each capability checkpoint since ChatGPT — reconstructed, and labelled as such.
—— this stretch contains a verified event- - - no event in this stretch — reconstruction only0 = no task exposed, 100 = every task exposed
● 4 verified events for this occupation, plotted at the date it happened — the parts of the line near a marker are anchored to something checkable.
The one curve on this site where the occupation barely existed at the left-hand edge, so the early points describe today's work projected backwards rather than a job anyone held in 2022 — worth saying plainly, because a reconstructed line looks equally confident either way. It is low and nearly flat because almost every task here is judgement inside one specific company: what to try first, what the tool may do unsupervised, why people went back to the old way. The gentle rise is evaluation tooling and code generation taking over parts of testing and integration. Read the low number as a description of the work, not a statement about the headcount: this function is funded as a transition cost, and the risk to it is the transition ending, which no automation index measures.
A flat line is not a forecast of safety. It says which tasks automation has reached so far — the occupations that moved least here are the ones where the constraint is physical or regulatory, and both of those can change.
Written about this#
These pieces argue from the same records this page holds, and each of their sections names what it rests on.
Method and sources#
- Assessment date
- 2026-09-11
- Basis of the task judgements
- 4 evidence-backed · 2 platform inference · 0 not enough evidence
- Verified events
- 4